The hidden cost of the undifferentiated approach
When all machines are assessed and treated equally, systemic inefficiencies silently erode competitiveness. On one side, valuable resources are wasted on excessive checks for marginal equipment. On the other hand, and this is the greater risk, business-critical machines may not receive the necessary attention until it is too late and an unexpected breakdown disrupts production.
Consider a concrete example. In a medical device company, the failure of a final assembly machine means:
- production losses
- potential customer complaints
- reputational damage
- legal consequences
In short: high costs.
Conversely, the failure of an auxiliary compressor can be managed calmly, perhaps by using a backup unit.
Applying the same level of predictive maintenance to both would mean wasting resources in the second case or risking disaster in the first.
A decision-making framework for the classification of machines
The solution lies in adopting a structured approach to machine classification, based on objective and measurable criteria. This framework, derived from the principles of Total Productive Maintenance (TPM), considers four key dimensions that together determine the real criticality of each asset:
- Safety as a non-negotiable foundation:
Before any economic consideration, a machine that poses risks to operator safety always requires maximum attention. Beyond regulatory compliance, corporate responsibility and operational continuity are at stake: a serious accident can stop production far longer than any mechanical failure. - Impact on product quality:
Some machines, if not properly calibrated or maintained, can generate defects that affect the entire value chain. In regulated sectors such as pharmaceuticals or food (and in luxury, where perfection is part of perceived value), this criterion can become predominant, a single non-conformity can compromise the output of an entire line. - Intensity and mode of use:
Attention must be paid not only to operating hours. A continuously running machine has different requirements than equipment used occasionally under optimal conditions. Moreover, some components wear based on specific parameters: number of cycles for presses, rotations for bearings, hours of high-temperature exposure for furnaces, or wear for high-tech assets. Understanding and evaluating these patterns is essential for accurate classification. - Systemic impact on the production chain:
Some machines represent real bottlenecks: if they stop, the entire flow halts. Others, instead, operate in parallel or have alternatives available. This dimension requires a deep understanding of the production layout and material flows, not just the individual machine.
The three maintenance strategies: allocating resources where they are needed
Analysis of the type of machine used and its impact on internal organisation, the product manufactured and the customer served, combined with the four criteria listed above, typically allows three categories of machines to be identified, each with its own optimal maintenance strategy.
Class A machines, which represent the beating heart of the company.
These are the ones whose downtime or malfunction would have immediate and serious consequences. For these, investment in advanced predictive technologies and preventive maintenance is not a luxury but a necessity. Sensors for vibration monitoring, thermographic analysis, wear control through oil analysis: these are investments that pay for themselves by preventing even a single critical breakdown. These systems require technology and specialist skills to interpret the data and take preventive action.
Class B machines, which form the operational backbone.
Important but not critical, they benefit from planned maintenance based on time intervals or usage parameters. Like servicing a car: regular interventions that maintain efficiency without excessive zeal. The key is to calibrate the intervals correctly, balancing the risk of failure with the cost of interventions.
Class C machines, which are support assets.
For these, the most efficient strategy may be incidental or “breakdown” maintenance, i.e. intervening when they break down. Investing in predictive systems for easily replaceable or little-used machines would in fact be a waste of resources that could be better used elsewhere.
The art of customization: every company is unique
The real value of this approach does not lie in the mechanical application of a formula, but in its ability to adapt to the specific needs of each production reality. The classification criteria must in fact reflect the strategic priorities of the company, without being derived from a pre-packaged scheme.
Let us take a few examples.
For a company competing on costs in a commodity market, production continuity and efficiency may be the main drivers. For a producer of luxury goods, where the unit margin is high, absolute quality and punctual delivery may prevail over considerations of production efficiency. In the fashion sector, with its rigid seasonal deadlines, a delayed delivery can mean the loss of an entire season, making any machine that could compromise timing a critical one.
Intervention thresholds must also be calibrated to the specific reality: defining as “critical” a machine that breaks down more than once a month may be unrealistic for a company with dated machinery, while it could be too permissive for a business with state-of-the-art equipment. The important point is to establish parameters that are both challenging and achievable.
Identifying and managing critical machines: the benefits of a differential approach
The implementation of a criticality classification system generates benefits that go far beyond the simple reduction of failures.
- optimization of resource allocation
Specialist skills, increasingly rare and costly, are concentrated where they generate the greatest value. The most experienced technicians focus on critical machines, while ordinary maintenance can be managed with standard resources. The aim is not to create hierarchies but to leverage skills where they really make the difference.
- intelligent reduction of total costs
Spending more on predictive maintenance for critical machines often leads to a reduction in overall costs, because the savings resulting from preventing critical stoppages far exceed the additional investments in selected equipment, while eliminating unnecessary interventions on non-critical machines frees up significant resources.
- greater operational predictability
With critical machines under predictive control and secondary ones managed appropriately, unplanned emergencies decrease drastically. This makes possible a more accurate production planning and smoother management of human resources.
- improvement of corporate culture
When the maintenance strategy is based on shared and objective criteria, collaboration between departments increases. Production and maintenance begin to speak the same language (and that is no small achievement!) based on data and clear priorities rather than on subjective perceptions or momentary emergencies.
The route to implementation: where to begin
The adoption of this approach does not require immediate revolutions.
The first step is to map the machine park and begin to collect objective data: frequency of failures, downtime, and production impacts. Even a simple spreadsheet may be sufficient to start. It is essential to have reliable data: often, seldom, sometimes, are adverbs and not KPIs on which to base a strategy.
In the second step, classification criteria specific to the production reality are defined. This stage requires the involvement of different corporate functions: production, maintenance, quality, safety. It is a process that takes time but builds knowledge, consensus, and shared understanding.
Finally, the implementation of a differentiated system can proceed gradually. It is not necessary to immediately equip all Class A machines with sophisticated and costly predictive systems: one can start with the most critical, demonstrate the benefits, and then progressively extend the approach where needed.
A strategic necessity, not an option
When efficiency, quality, and reliability increasingly make the difference, the diversified management of critical machines is no longer a nice-to-have but a strategic necessity. Companies that continue to apply undifferentiated approaches to maintenance find themselves caught between two fires:
- waste of resources on non-critical assets
- unacceptable risks on fundamental machines
The question is therefore not whether to implement a criticality classification system but how quickly it can be done. Because while you are reading this article, somewhere a critical machine not yet identified is approaching failure, and when it happens, the cost will be far higher than that of any predictive system you could have implemented.
True operational excellence does not lie in doing everything perfectly, but in doing perfectly what really matters. And to do so, you must first know what really matters.”
Virginio Peluzzi – Partner ŌdeXa
ŌdeXa supports companies in optimizing maintenance strategies through the identification and management of critical machines, with a pragmatic approach oriented towards measurable results and the economic sustainability of interventions.
